Private returns to higher education in NigeriaThus, the higher the level of education, the higher the rate of return to the individual. Efforts should be made to improve the quality

Private returns to highereducation in Nigeria

By

O.B. OkuwaDevelopment Policy Centre, Nigeria

AERC Research Paper 139African Economic Research Consortium, Nairobi

March 2004

© 2004, African Economic Research Consortium.

Published by: The African Economic Research ConsortiumP.O. Box 62882-00200Nairobi, Kenya

Printed by: The Regal Press Kenya, Ltd.P.O. Box 46116-00100Nairobi, Kenya

ISBN 9966-944-33-8

ContentsList of tablesList of figuresAcknowledgementsAbstract

1. Introduction 1

2. Background of the study 2

3. Literature review 9

4. The model 15

5. Empirical analysis 19

6. Summary 26

References 29

List of tables

1: Distribution of student enrolment and number of schoolsby type of education institutions and academicYear 1983/84 – 1994/95 6

2: Distribution of graduate turn-out and number of schoolsby type of educational institution and academicYear (1983/84 – 1994/95) 6

3: Characteristics of employed graduates of education institutions 74: Mean monthly earnings by educational level by sex and sector (1995) 195: Monthly earnings differentials associated with schooling 206: Means earnings by experience group, sex and sector 217: Private rate of returns to schooling (all workers) 228: Coefficient on education dummies (Public Sector Workers) 239: Coefficient on education dummies (Private Sector Workers) 2310: Private rate of return to an additional year of education (%) 24

List of figures

1: Possible channels to higher education 162: Alternative routes to higher education 16

Acknowledgements

I am highly indebted to the African Economic Research Consortium (AERC) for providingfunds, materials and technical assistance through the useful suggestions and commentsof Groups A and B resource persons: Prof. A.A. Ali, Prof. David Sahn, Prof. GermanoMwabu, Dr. Dominique Njinkeu, Dr. Mohsin Khan, Dr. Shanta Devarajan, Prof. SteveO’Connell and Dr. Nadeem Ul Haque. I wish to thank the research fellows in the samegroup for their contribution to the improvement of this paper. Finally, I am grateful toMr. Meshack Aziakpono and Mr. Dele Akinpelu for their contributions. However, I takefull responsibility for the contents of the paper.

Abstract

The idea of education as a capital good is rooted in the concept of “human capital”,which attaches a high premium to human skills as a factor of production in the developmentprocess. Education is known to be an important determinant of earnings in marketeconomies. The higher an individual’s educational attainment, the higher that individual’sexpected starting salary and the steeper the rise in earning capacity over time, especiallyduring the early working years. This study investigates the variation in the rate of returnto different levels of education.

The study endeavours to determine the relationship between years of schooling andearnings (rate of return) in Nigeria. The effect of the amount of time in the labour forceon earnings is also examined and an attempt made to highlight and calculate the privaterates of return to graduates of higher educational institutions according to sectors of theeconomy and sex of the graduate.

The descriptive statistics and ordinary least squares (OLS) estimation results revealthat the mean monthly earnings of workers increase with more years of schooling. Theprivate rate of returns is low for graduates of colleges of education. It is higher forpolytechnic graduates and higher still for university graduates. Furthermore, the meanearnings increase with higher years of labour market experience.

Thus, the higher the level of education, the higher the rate of return to the individual.Efforts should be made to improve the quality and investment in this level of educationby encouraging private individuals to invest in and pay for higher education.

PRIVATE RETURNS TO HIGHER EDUCATION IN NIGERIA 1

1

1. Introduction

A major controversy among analysts and policy makers concerns the objectivesof educational development. Some have suggested that education should beprovided for its own sake, as a means of enriching individuals’ knowledge and

developing their full personality. This concept of education has continually influencedpolicies in some advanced countries of the world. Others hold that education shouldseek to prepare people to perform functions that are essential for the transformation oftheir environment. The two points of view can be considered in terms of regardingeducation as a consumer good or as a capital good. Nigeria should, in its present stage ofdevelopment, regard education as both (Second National Development Plan, FederalGovernment of Nigeria, 1970–1974).

The notion of education as a capital good is rooted in the concept of “human capital”,which attaches a high premium to human skills as a factor of production in the developmentprocess. A corollary of this is that human skill or productivity is just as important aninput in the process of development as finance, natural wealth and physical plant. Becauseeducation plays a most important role in the creation and improvement of human capital,its relevance and importance to economic growth and development are now very wellrecognized in development planning. Experiences of developing countries during thepast decades have indicated that shortage of talents and skills needed for developmentcan decisively retard economic progress (World Bank, 1995).

Therefore, a country like Nigeria cannot afford to leave education to the whims andcaprices of individual choice. Since available resources for development are highly limited,public policies in the field of education must take full account of the needs of the countryin terms of the development of manpower and skills.

2 RESEARCH PAPER 140

2. Background of the study

I n Nigeria, the decline in the quality of education at all levels has become a fact ofnational life. Indeed, the most significant event in the sector in the recent past hasbeen the continuing crisis that besets the educational system. This crisis is rooted in

the deteriorating conditions within the citadels of learning, in respect of teaching facilitiesand other infrastructural facilities, the welfare of those engaged in the teaching professionand the ever increasing cost of education. This has culminated in student strikes andindustrial actions by teachers at all levels of the educational system. As observed in theThird National Development Plan, at the primary level the shared responsibility of statesand local governments in managing the schools leaves neither of the tiers of governmentresponsible for the upkeep of the system. The result is that infrastructural facilities arenot maintained and teachers’ salaries are not paid for months. The same is applicable atthe secondary level even though state governments have the sole responsibility for thattier of the educational system. At the tertiary level, the facilities are also rapidlydeteriorating.

Several attempts have been made by government to reverse the deteriorating trend inthe educational system. For example, in November 1990, the federal governmentconstituted the Longe Commission on the Review of Higher Education in Nigeria with aview to redressing the situation. The recommendations of the Commission were largelyleft unimplemented by 1993, which gave rise to an industrial action declared by theAcademic Staff Union of Universities (ASUU) that lasted for over five months. Sincethen, industrial actions by ASUU and Nigerian Union of Teachers have become a yearlyevent. This has disrupted the academic programme at virtually all levels.

Statement of the problem

Given the rising cost of education, government took steps to improve the availabilityof resources to prosecute its education programme by enacting Education Tax Decree

7 of 1993. The decree provided for 2% of the accessible profit of a company registeredin Nigeria to be collected by the Federal Board of Inland Revenue and paid into a fundknown as the Education Fund. The fund, which was to be managed by the Education TaxBoard of Trustees, was to be disbursed to federal, state and local government educationinstitutions, principally for work centres, staff development, conference attendance, librarysystems at different levels of education, research equipment procurement, maintenanceand purchase of higher education books.

2


Owing to the failure of the state and local governments to fund primary educationappropriately, the federal government moved to take over the affairs of that tier of thesystem. Another major development in funding education in recent years was theestablishment of the Education Bank. The bank is to take over the responsibilities of thedefunct Student Loans Board in assisting Nigerians who might have limited access toeducation.

Despite these efforts by the governments, the crisis in the Nigerian educational system,particularly the tertiary level – universities, polytechnics and colleges of education – andtheir fundamental causes, that is, the gross under funding of the institutions, poorconditions of service of the academic staff among other issues, has continued unabated.It is clear that tremendous resources, both human and material, are required to redressthe situation. Considering the enormous cost of running education in this country, it isgenerally advocated among policy makers that education funding should be the jointresponsibility of the government, the private sector and the beneficiaries. The argumentthat beneficiaries of education, especially higher education, be made to pay at least aportion of the cost of educating themselves is anchored in the principle of moderateprivatization or partial cost recovery or some kind of selective pricing. According toJimenez and Tan (1991), this policy recommendation centres on the belief that returns(private) to individuals of education increase the higher the level of educational attainment(Psacharopolous, 1994).

There is an enormous literature devoted to estimating rates of return to schooling andon-the-job training, but there are few empirical studies from sub-Saharan Africa, andnone from Nigeria, so far as the author is aware. This study helps to fill this vacuum byusing data generated from a recent labour market survey by National Manpower Boardin collaboration with the National Population Commission in 1995.

Objectives of the study

The study is basically designed to evaluate the nature of returns to higher educationi n

Nigeria with a view to accounting for the variation in the rate of returns to differentlevels of education. Specifically, the study has the following objectives:• Determine the relationship between years of schooling and earnings (rate of return).• Assess the effect of the amount of time in the labour force (experience) on earnings.• Highlight and calculate the private rates of return to graduates of higher educational

institutions according to sectors of the economy and sexes.• Offer policy recommendations that will help to enhance the development of education

in Nigeria.

Research questions and hypothesis

I n an attempt to achieve the stated objectives, the study addresses four basic questions:

• Do more years of schooling increase the earnings of an individual?


• Do the individual’s sex and the sector of work affect earnings?• What impact does work experience have on an individual’s earning?• What policies emanating from the findings can help improve the development of

education in Nigeria?

Our hypotheses are:• That there is a positive relationship between the wages of an individual, the years of

schooling and the individual’s working experience.• That earnings vary across sectors and between sexes in Nigeria.• That as bad as the education system is at present, it can be improved substantially

over a relatively short time if the enabling policies are put in place.

Justification for the study

Despite the importance of and need for education, many low-income countries(African) still give it less attention and lack appropriate policies to promote

educational expansion. Policies are often adopted without due consideration for theparticular nature of the developing countries. And because of the declining economicoutput in the African continent, coupled with economic and political instability, mosteducational institutions have been subjected to serious under-funding. In view of theprevalent situation, from which Nigeria is not excluded, there is an urgent need for wellarticulated and well informed policies to be put in place to address the problems ofeducational underdevelopment. It is in this direction that this study is conceived andcarried out.

Profile of higher education in Nigeria

Formal education in Nigeria started as a private sector affair and extended to theworks of the missionaries in the colonial era. The government took interest in

education quite early and in 1872 the Lagos administration gave a grant to the missionarysocieties to provide education in the colony of Lagos.

Higher education has been a fast growing segment of the Nigerian educational systemduring the last decade. Its development dates back to the decision reached at the Asquithand Elliot Commission of 1943. The development of higher education is correlated witheconomic development, however. The term “higher education” encompasses variousforms of educational institutions beyond the secondary school level. These include theconventional universities, which offer courses in both the sciences and humanities, andspecial universities for sciences, agriculture or engineering. It also includes polytechnics,which provide advanced vocational training, professional schools such as managementor public administration schools, and the colleges of education, which train professionalteachers.

There has been a rapid expansion in post secondary education facilities in the countryover the 37 years since independence. The number of universities has risen from 1 to 37,


of which 21 are controlled by the federal government while 9 are state owned.Out of the21 federal universities, 5 are universities of technology and 2 are universities of agriculture.

As at the 1989/90 academic session, student enrolments in the universities stood at172,911; by 1994/95 this had increased to 261,780, compared with 3,800 in the 1962/63session. Similar expansion has taken place in the polytechnics and colleges of education.

The number of polytechnics grew from 3 to 36 between 1960 and 1995, while collegesof education, which started as Federal Advanced Teachers Colleges, grew from 30 to 61within the 15-year period from 1980 to 1995.

Student enrolment in the polytechnics stood at 75,468 at the 1989/90 academic sessionand increased to 140,953 by the 1994/95 academic session. For the same periods, 1989/90 and 1994/95, the enrolments in colleges of education were 72,525 and 108,373,respectively. Table 1 shows the distribution of student enrolments by type of educationalinstitution.

Ten years after independence, the universities turned out a total of 2,623 graduates, afigure that rose to 38,367 in the 1988/90 academic session. It increased further to 48,219by the end of the 1994/95 academic session. The expansion in graduate turnout in thepolytechnics was not as rapid as that of universities; by the end of the 1989/90 academicsession it was 28,656, which increased to 43,965 by the end of the 1994/95 academicsession. Table 2 shows the distribution of graduate turnout and number of schools bytype of educational institution. Universities take the lead followed by polytechnics andcolleges of education.

The desire of people to obtain a degree from any of these three types of higherinstitutions could be social, economic or political. The earnings of individuals (or privatereturns) that accrue to education are part of the reasons for attaining this higher level ofeducation. The extent or degree of association between returns and educational attainmentwill become apparent later in this paper.

The data for this study are drawn from a Nigerian labour market survey by the NationalManpower Board in collaboration with the National Population Commission in October1995, which represents the most recent and comprehensive data on labour marketcharacteristics. The survey covered only Lagos state, but included a randomly selected302 of the 12,145 enumeration areas (EAs) in the state. The EAs were produced from the1991 National Population Census and included 24,737 households. The sample yielded3,187 households and 14,192 respondents.

Though the coverage is relatively small, it represents the best that could be obtainedat present. Respondents from all sectors of the economy – public and private,manufacturing and non manufacturing are represented. The survey yielded informationon earnings, age, sex, marital status, highest educational qualifications, specializededucation, employment experiences and other personal characteristics. Given the focusof this study (analysing private returns to higher education in Nigeria), only relevantdata were extracted from the data set.

As a first step, we provide the characteristics of respondents that have all the requiredinformation in Table 3. Of the 2,053 employed respondents, 1,078 respondents weregraduates of higher institutions, while 975 respondents were graduates of lower levels ofeducation. The set of the sample that graduated from higher education institutions yieldeda sex ratio of 368 males to 171 females as shown in Table 3a.

6 RESEARCH PAPER 140Ta

ble

1: D

istr

ibut

ion

of s

tude

nt e

nrol

men

t and

num

ber o

f sch

ools

by

type

of e

duca

tiona

l ins

titut

ions

and

aca

dem

ic y

ear 1

983/

84 –

199

4/95

Year

/In

stitu

tion

1983

/84

1984

/85

1985

/86

1986

/87

1987

/88

1988

/89

1989

/90

1990

/91

1991

/92

1992

/93

1993

/94

1994

/95

Uni

vers

ityN

o. s

choo

ls24

2428

2830

3032

3234

3537

37

Enr

olm

ent

108,

753

123,

743

133,

626

148,

720

158,

757

164,

004

172,

911

195,

759

222,

974

233,

7467

246,

265

261,

780

Pol

ytec

hnic

No.

sch

ools

2727

2727

2828

2831

3234

3636

Enr

olm

ent

61,2

0062

,787

64,9

3365

,905

68,6

7572

,134

75,4

6810

6,92

611

1,80

612

1,52

713

1,97

814

0,95

3

Col

lege

of

Edu

catio

nN

o. s

choo

ls41

4242

4448

5051

5461

6161

61

Enr

olm

ent

51,0

2057

,443

57,4

8164

,818

41,8

9067

,757

72,5

2585

,574

92,3

9397

,836

103,

706

108,

373

Sou

rce:

Nat

iona

l Pro

duct

ivity

Cen

tre

Pub

licat

ion,

199

4.

Tabl

e 2:

Dis

trib

utio

n of

gra

duat

e tu

rn-o

ut a

nd n

umbe

r of s

choo

ls b

y ty

pe o

f edu

catio

nal i

nstit

utio

n an

d ac

adem

ic y

ear (

1983

/84

– 1

994/

95)

Year

/In

stitu

tion

1983

/84

1984

/85

1985

/86

1986

/87

1987

/88

1988

/89

1989

/90

1990

/91

1991

/92

1992

/93

1993

/94

1994

/95

Uni

vers

ities

No.

sch

ools

2424

2828

3030

3232

3435

3737

Turn

out

25,8

2227

,500

30,4

8931

,864

37,2

8638

,367

40,0

9441

,497

42,9

0844

,624

46,4

5448

,219

Pol

ytec

hnic

sN

o. s

choo

ls27

2727

2728

2828

3132

3436

36

Turn

out

11,8

6820

,946

25,9

5922

,426

25,5

7827

,450

28,6

5631

,321

34,2

8437

,418

40,9

8943

,905

Col

lege

s of

edu

catio

nN

o. s

choo

ls41

4242

4448

5051

5461

6161

61

Turn

out

13,4

3216

,432

18,2

5519

,049

19,8

0320

,374

21,0

9521

,757

22,4

4023

,114

23,8

7025

,684

Sou

rce:

Nat

iona

l Pro

duct

ivity

Cen

tre

Pub

licat

ion,

199

4.


Males accounted for 68.3% and females 31.7% of the total sampled population. Theeducational qualifications of the sampled population were of the order of 20.3%, 28.5%and 51.2% for colleges of education, polytechnics and university education. The majorityof the graduates were employed in the public sector (44.5%), followed by the privatesector (31.1%), while the self-employed were fewest at 24.4% of the sampled population.

At the lower educational level, the sample yielded a sex ratio of 67.3 male to 32.7female. A majority of the lower education graduates are employed in the private sector(57.3%), while the public and self-employed sectors recorded 42.7% and 18.5%,respectively.

Table 3: Characteristics of employed graduates of education institutions

a) Graduates of higher education institutions

Variable Means Proportions (%) Standard deviation

Age (years) - 37.8 9.46Earnings (N) - 6982.7 ($87.3) 950.67Work experience - 17.32 9.46Male 68.3 - -Female 31.7 - -Public 44.5 - -Private 31.1 - -Self-employed 24.4 - -NCE 20.3 - -Polytechnic 28.5 - -University 51.2 - -No. respondents 1078 - -

Note: $1 = N80

Source: Computed by author, based on the Nigerian Labour Market Survey by National Manpower Board in1995.

b) Graduates of lower education

Variable Means Proportions (%) Standard deviation

Age (years) - 34/26 10.17Earnings (N) - 3017.21 ($38.1) 4565.85Work experience - 24.47 10.73Male 67.3 - -Female 32.7 - -Public 42.7 - -Private 18.5 - -Self-employed 23.6 - -Primary 76.4 - -Secondary - -No. Respondents 975 - -

Note: $1 = N80

Source: Computed by author, based on the Nigerian Labour Market Survey by National Manpower Board in1995.


Information on means and standard deviations of age, work experience and earningsof the sampled population are also provided. Experience was approximated by age minusyears of education and sex, which is the standard procedure for this type of analysiswhere precise data are not available (Cohen and House, 1994). The proportions of workexperience in the sample are 17.32 and 24.47, respectively, while the mean ages are 37.8and 34.26 for higher and lower education. On the other hand, the mean monthly earningsamounted to N6,982.82 and N3,017.21.

However, the earnings cannot be assumed to be the same among educational andexperience groups, among the sectors, whether public or private, or industrial sectors,agriculture, mining, manufacturing and so on. Similarly, as often argued in the literature,differences in earnings between the sexes are expected. These arguments will be pursuedin a subsequent section. Particular attention is focused on the contributions of differencesin schooling as well as workers’ experience to earning differentials.


3. Literature review

The importance of education for economic growth and development and its expectedreturns to individuals, as well as the society at large, has attracted great interest inliterature in both developed and developing countries. The growth in both

theoretical and empirical literature on education in the last two decades is not unconnectedwith the increasing importance being attached to education in the process of economicdevelopment. At the risk of being all embracing, attention is focused on literature onreturns to educational investment. In particular, the concern is with the methodologicalissues surrounding the estimation of rates of return and the empirical pattern of rate ofreturn estimates.

Methodological issues

To begin with, the rate of return to educational investment can be private or social.Todaro (1982) defines private rate of return as the gains that accrue to a single

individual from attaining a particular level of education, whereas social returns refer tothe gains or benefits that accrue or are available to the society as a whole. According toPsacharopolous (1994), private rates of return are used to explain people’s behaviour inseeking different educational levels and types, and as distributive measures of the use ofpublic resources. Social rates of return can be used to set priorities for future educationalinvestments. But how can these be estimated?

Three distinct approaches can be distinguished in the literature for providing estimatesof the profitability of investment in education. These three approaches in order ofincreasing complexity are:1. The average earnings by educational level;2. The “earning functions” method, which has two variants (Psacharopolous 1994);

and3. The “full” or “elaborate” method (Grindling et al., 1995).

We briefly highlight these approaches in turn. For a fuller discussion of the differentrate of return estimation methods, see Psacharopolous and Ng (1994). Suffice to say thatthe method adopted by various authors is often dictated by the nature of the availabledata.

The first and simplest estimates of the returns to education use descriptive statisticsthrough the calculation of percentage differences in mean wages between each education

9


group (like primary graduates versus secondary graduates). This approach provides afirst approximation to the rate of return to education at different levels (Grindling, Getal., 1995). However, the argument that some of the differences in average wages betweeneducation levels could be due to differences between the workers at each educationallevel in other respects – determining characteristics like experience, age and the like –calls for other approaches. The second approach attempts to address this problem byestimating an earnings equation with a variety of controls for other earnings determiningcharacteristics as well as dummy variables for each educational level.

The “basic” earnings functions method is due to Mincer (1974) and involves thefittings of a semi-log ordinary least squares (OLS) regression using the natural logarithmof earnings as the dependent variable, and years of labour market experience and itssquare as independent variables. In this semi-log earnings functions specification, thecoefficient on years of schooling can be interpreted as the average private rate of returnto one additional year of education, regardless of the educational level to which this yearof schooling refers (Psacharopolous, 1994; Grindling et al., 1995; Cohen and House,1994; Psacharopolous and Ng, 1994).

Another variant of the earnings function involves regressing the dependent variable(wage rate) on education and experience again, but this time the education is brokendown into a set of dummy variables representing different educational levels. In thisversion, the coefficients on the variables for education are often interpreted as returns tothe level of education (Cohen and House, 1994). As noted by Chiswick (1997), whilethis may sometimes seem to be a correct interpretation, in principle and in manycircumstances it is not. Hence some authors move a step further in order to obtain theestimated rate of return to an additional year of schooling by dividing the differencebetween the coefficients of adjacent groups by their differences in years of schooling(see Cohen and House, 1994; Psacharopolous and Ng, 1994).

The elaborate or full method follows from the exact algebraic definition of the rate ofreturn. The method amounts to working with detailed age–earnings profiles by level ofeducation and finding the discount rate that equates a stream of education benefits to astream of education costs at a given point in time (Psacharopolous, 1994). Though thismethod is regarded as the most appropriate (among those listed above), because of itsenormous data requirements, researchers have resorted to less data-demanding methods.Indeed, authors have found it increasingly convenient to estimate the returns to educationon the basis of the Mincerian earnings functions method. In theory, and as many empiricalstudies have shown, the earnings functions and the elaborate methods should give verysimilar results (see Psacharopolous, 1994, for the results of studies using these twoapproaches).

Some recent empirical studies on returns to education have added another dimensionto the methodological approaches discussed so far – the problem of sample selectivity.The possibility of sample selection bias arises whenever one examines a subsample, andthe unobservable factors determining inclusion in the subsample are correlated, with theunobservables influencing the variable of primary interest (Vella, 1997).

In the literature, several remedies exist. The first solution to this problem was suggestedby Heckman (1979), who proposed a maximum likelihood assumption regarding Œ

i u

i).


Other approaches that have found more frequent application in the literature are the two-step estimators. These can be categorized into three groups. The first fully exploits theparametric assumptions in at least one stage of estimation. The second is semi-parametricin that it relaxes the distributional assumptions, and the third focuses on conditionalexpectations and bounds. It is not our intention here to provide a detailed review of thesemethods, as these have been thoroughly done in other studies (Vella, 1997).

Considerable attention has been given to testing and correcting for selectivity biasproblems in rates of return to education studies in the world using these approaches.They have found little application in African studies, however. The dearth of data requiredfor implementing these models has continued to be a major constraint. For example, as astarting point in testing and controlling for selectivity bias, one needs to estimate a labourforce participation equation that is a function of own and household characteristics,parental characteristics, human capital variables, ethnicity, and a variable for place ofbirth (Krishnan, 1994). Own and household characteristics include age, marital status,the number of very young (below five years of age) or old (above 65 years of age)dependents, total family size, whether head of the household, whether migrated recently,and ethnicity. In most African countries (Nigeria inclusive) where labour market dataare still at the primitive stage, most such data requirements remain a luxury.

Empirical Issues

I t is perhaps convenient to commence discussion with the work of Psacharopolous(1994), which provides a comprehensive review of recent literature on returns to

education investment. Several dimensions of returns to education were highlighted. Thereview covered 20 studies that used the full method and the Mincerian rate of return for78 and 62 countries, respectively. Central to most of the studies was the investigation ofthe issue of the indisputable and universal positive correlation between education andearnings. However, the interpretation is varied and often conflicting. The major findingsof Psacharopolous (1994) are summarized below.

First, he observed that among the three main levels of education, primary educationcontinues to exhibit the highest social profitability in all world regions. Furthermore,private returns are considerably higher than social returns, a situation he attributed to thepublic subsidization of education. Second, he observed a declining pattern of the returnsto education over time, where all social returns declined between two and eight percentagepoints on average in a 15-year period. However, he noted that the returns to highereducation increased by about two percentage points during this period.

When gender consideration was examined, his finding confirms that overall the returnsto female education are higher than those for males, although individual levels of educationshow a more mixed pattern. Even when estimates were adjusted for selectivity bias, thatis, by taking into account the prior decision of a woman on whether to participate in thelabour force (Heckman, 1979), the rate of return estimate for females remained virtuallyunaffected, and the returns experienced by females, whether corrected or not, exceededthose for males by more than one percentage point. Moreover, the review shows a large


variation between the returns to higher education faculties, the lowest social returnsbeing for physics, sciences and agronomy, and the highest private returns for engineering,law and economics. Similarly, the sector of employment accounted for some differencesin returns, with returns in the private/competitive sector of the economy higher thanthose in the public/non-competitive sector. Likewise, he observed that the returns in theself-employment sector of the economy are somehow lower than in dependentemployment.

A major fact emerging from the findings of Psacharopolous (1994) is that variation inearnings or returns to education is a function of many factors and not only the years ofschooling. Among the variables that have been included in the debate are experience(age minus years of formal education minus six), quality of education and socio-economicbackground. A brief review of the evidence on these variables should at least serve as acaveat in the interpretation of the coefficient of years of schooling.

On the interactions among education, earnings and ability, Chou and Lavin (1987)introduced progressive matrixes as proxies for genetic ability in an agricultural productionfunction in Thailand and found that the effect of education on farm productivity (earning)is upheld. Psacharopolous and Veloz (1992), in a study in Colombia, introduced reasoningability (measured by means of matrixes) and the coefficient of years of schooling wasreduced from 10.5 to 9.4%. Moreover, Glewwe (1991), using the matrixes variable in anearnings function in Ghana, failed to register an effect different from zero in the earningsdetermination process.

The issue of the returns to investment in the quality rather than quantity of educationhas come to be a frontier of research in this field. Card and Krueger (1992a) examinedthe effect of school quality on the returns to education using 1980 US census data.Measuring quality by the student–teacher ratio, the average term length and the relativepay of teachers, they found that people educated in states with high quality schoolsexhibit higher returns to additional years of schooling. For example, a decrease in classsize from 30 to 25 pupils per teacher is associated with a 0.4 percentage point increase inthe return to education.Regarding the role of socio-economic background, Card andKrueger (1992b) found that holding school quality constant, there is no evidence thatparental income or education affects state-level returns to education. In another paper,Card and Krueger (1992b) found that improvements in quality of education that blacksreceive explain 20% of the narrowing of the black and white earning gap in the UnitedStates during 1960 to 1980.

Another debated issue in the literature has been the role of soci-economic background.Card and Krueger (1992a) found that, holding school quality constant, there is no evidencethat parental income or education affects state-level returns to education. But Newman(1991), using Israeli data found that the returns to schooling are higher for those comingfrom favourable socio-economic background.

The results obtained by Grindling et al. (1995) represent a significant divergencefrom the findings of Psacharopolous (1994). Using the three methods of estimating returnsto education and data from the May labour force survey of the Taiwan area from 1978 to1991, they obtain some striking results. First, they found that private rates of return inTaiwan are highest for higher education levels (for example, university) and lowest for


lower education levels (for example, junior high school). This is quite in contrast to theresults observed in Psacharopolous (1994). Similarly, unlike most other studies ofchanging returns to education over time in developing countries (see Schultz, 1963) andPsacharopolous’ (1994) review, they found that private returns are higher for men thanwomen at all educational levels.

The study by Hossain (1976) deserves some attention at this juncture. Using theelaborate and earnings function methodologies, the study estimated both social and privaterates of return to three levels of education in China. Using data from a 1993 laboursurvey by the Ministry of labour, the study obtained high returns (private and social) toeach level of education in 1993. The social rates of return were highest for primaryeducation at 14.4%, followed by 12.9% for secondary education and 11.3% for highereducation. While the private rates of return were also highest for primary education(18.0%), this was followed by higher education and secondary, respectively. This patternof rate of return (private and social) was attributed partly to higher government subsidyin terms of increased operating budget and teachers salaries.

The study by Cohen and House (1994) examined the relevance of the human capitalapproach to explaining the variance in workers’ productivity and earnings in the labourmarket of urban Khartoum. The findings of the study tend to lend credence to thecontroversy in the literature on returns to different levels of education. One of theirprincipal findings is that returns to primary education are lower than the average forother developing countries, while returns to college education are higher. The resultsoppose the popular view observed by Psacharopolous (1994).

The World Bank (1995) study on Vietnam education financing provides support forthe thesis that primary education generates greater returns than secondary and tertiaryeducation. In an attempt to reach a fairly robust conclusion, several different methodswere used to assess the benefits of education and training relative to the costs of investingin these programmes. The general picture, drawing on different methods and the mostrecent information available, is that rates of return to education and training in Vietnamare low by international standards. For example, the rate of return is only about 3%higher for some levels of education and for some categories of workers, but lower forothers. The results suggest that in Vietnam in the early 1990s most education and traininginvestments were marginal investments. Primary education satisfied the 10% rate ofreturn test, given the assumptions of just one year’s forgone earnings, but secondary andtertiary education did not, although the private rate of return was higher for tertiary thanfor secondary. The findings also reveal that the rate of return was higher for investmentsmade in girls’ schooling than for boys, and higher for the education of those individualswho ended up working in the public sector, provided that their education did not stopafter primary school.

In the case of Nigeria, studies are very scarce on returns to education investment. Inwhat could perhaps be regarded as the pioneer study in this area, Psacharopolous (1985)used data obtained from a 1966 survey to estimate both social and private returns forprimary, secondary and higher education using the full method. Basically the study obtainshigher rates of returns (private and social) to each level of education for 1966, with theprivate returns higher than the social returns. The social rate of return was highest for


primary education (23.0%) followed by higher education (17.5%) and secondary education(12.8%). On the other hand, the private rate of return was highest for higher education(34.0%), followed by primary education and secondary education with returns of 30.0%and 14.0%, respectively. The high level of private returns to higher education in Nigeriaas found in the study is quite in contrast to the popular view noted in Psacharopolous(1994).

Three main conclusions are discernible from this review. First, the patterns of returnsto education at different levels remain inconclusive. Second, it appears from all thestudies that higher or tertiary education is assumed to be synonymous with universityeducation. The implicit assumption is that returns are equal for all higher education. InNigeria, as in many other countries, higher education comprises colleges of education,polytechnics and universities, all with different types of training and degrees/certificatesawarded. It may be unrealistic to assume that returns are equal across these classes ofhigher education. Third, the results obtained are not significantly sensitive to the methodsused for estimating the returns, whether Mincerian earning functions or full methods.


4. The model

Two distinct analytical methodologies are used in this study:descriptive statisticsand the modified Mincerian earnings function. Descriptive statistics are providedto highlight the earnings of graduates of educational institutions according to

sectors of the economy and sexes. Simple averages, standard deviations, and percentagedifferences in mean wages between each education group, sex and sectors, among others,are calculated. As a first approximation, the descriptive statistics give us insight into therelationship between years of schooling and earnings and the effect of labour experienceon earnings.

Analytical framework

I n order to increase our understanding of the relationship between schooling andearnings, we specified and estimated a modified Mincerian earnings function by

regressing the natural logarithm of the monthly wage rate (Lny),on education and

experience, with the education broken into a set of dummy variables representing differenteducational levels. The model is specified thus:

Lny = ∝

o + ∝

3CoE + ∝

4Pol + ∝

3Uni + ∝

6Ex + ∝

7Ex2 + E (1)

where:L

ny = the natural logarithm of the monthly wage rate

CoE = dummy for college of education graduatePol = dummy for polytechnic graduateUni = dummy for university graduateEx = labour market experienceEx2 = square of labour market experienceE = stochastic error terms

Two fundamental assumptions are made to facilitate our analysis about the processof higher education levels. In the first instance, we assume that there is a sequence ofhigher educational levels. Three possible channels are considered, as illustrated in Figure1.

15


Figure 1: Possible channels to higher education

PRY = primary SEC = secondary UNI = universityPOLY = polytechnic NCE = national colleges of education

The first channel (A) assumes that a graduate went through NCE and then to theuniversity. The other two channels are the more common ones. This assumption ispredicated on experience from labour markets and the years of schooling involved ineach of the higher education levels, which place a higher value on university, polytechnicsand NCE, in that order.

Second, and as an alternative, we assume that other higher levels of education –Nigerian colleges of education and the polytechnics – are alternative routes aftersecondary school education. Figure 2 illustrates.

Figure 2: Alternative routes to higher education

PRY = primary SEC = secondary POLY = polytechnicNCE = national colleges of education

UNI (A)

UNI (B)

UNI (C)

NCE

SEC

POLY(ordinary National DiplomaCertificate)

PRY

NCE (D)

POLY (E)

SECPRY


With the first assumption, we are able to analyse the returns to an additional year ofschooling among the higher education levels. The second proposition enables us toexamine the returns to higher education in relation to lower educational level, and thencompares the returns of the three types of higher education.

Out of these five possible channels of educational attainment, the most prevalentroute is B, i.e., PRY › SEC › UNI. This is so because this route takes a shorter time toachieve an educational goal and costs less than any of the other routes.

Channel B is followed by channels E, D and A. Most students who finish secondaryschool education and enter a polytechnic (for Ordinary National Diploma) or college ofeducation do not stop at these levels. They further their education by seeking admissioninto the university for a higher degree. The prevalence of route B, followed by E, D andA, is supported by the number of students enrolled in these higher institutions (Table 1).

Following these assumptions therefore, Equation 1 was re-specified as follows:

In Y = ∝o + ∝

1Pry + ∝

2Sec + ∝

3CoE + ∝

4Poly + ∝

5Uni + ∝

6Ex + ∝

7Ex2 + E (2)

InY = CoE, Poly, Uni, Ex and Ex2 are as defined above.Pry dummy – for primary education graduateSec dummy – for secondary education graduateE – the stochastic error terms

Note that the influence of schooling is modelled here as separable from the influence ofexperience; µ

o represents the entry-level wage to a new labour market entrance with a

lower education (compared with the one in the model) or no schooling, µ1, µ

2, µ

3, µ

4 and

µ5 are the coefficients of the dummies for primary, secondary, college of education,

polytechnic and university, respectively, which capture the marginal wage effects andare used to compute the return to their level of education. While µ

6 and µ

7 are intended

to capture returns to on-the-job training (experience), which is assumed to be non-linearbecause of diminishing marginal returns to increased on-the-job training and risingmarginal cost of further training over time. It is expected that:

∝0,

∝1, ∝

2, ∝

3, ∝

4, ∝

5, ∝

6 > 0 and ∝

7 < 0

α 7 0<The estimated rate of return to an additional year of schooling is obtained by dividing

the difference between the coefficients of adjacent groups by their differences in yearsof schooling. To arrive at these rates of returns, we concentrate on Equation 2, thus:

RS Spry

secsec

= −−

α α2 1 (3)


RS Scoe

coe

= −−

α α3 2

sec (4)

RS Spoly

pol

= −−

α α4 2

sec(5)

RS SUni

uni

= −−

α α5 2

sec(6)

where:S = number of years of schooling of the subscripted educational level.

A brief comment on this approach is necessary here. The model described above hasits root in the work of Mincer (1974), and it has become the dominant procedure inestimating private rates of return to education. The use of the dummy variable methodrather than the years of schooling squared method adds a great deal of sensitivity to theresult of private rates of return. This approach has generated much debate in the literature.

However, many of the issues can be traced back to the validity of an initial set ofsimplifying assumptions that were introduced to level mathematical tractability to theproblem. These issues include the following:• There is no control in the model for the quality of schooling.• All results refer only to wage employees.• The effects of schooling and experience should not be regarded as weakly separable.• The market should not be assumed to be in long-run equilibrium.• There are no corrections for unobserved ability bias that is correlated with school

attainment.• The amount of schooling may be measured with errors.• There are no controls for background variables such as parental education (Cohen

and House, 1994).

As is often the case in the social sciences, the theoretical debate is advanced to apoint where the requirements for adequately testing such models are considerable andfar exceed the available data. This is certainly true in Nigeria and elsewhere in sub-Saharan Africa where many micro-level data on employment and earnings are rare. Giventhe preliminary nature of this analysis and the limited scope of available data, it wasdecided initially to retreat from the intractable problem of attempting to control forunobserved ability and some other issues generally raised in the debates and to follow awell-established empirical literature by estimating Equation 2 using ordinary least squares(OLS) and then use the results to compute the rate of return using equations 3 to 6. Themodel was estimated for all workers, men only, women only, private sector only andpublic sector only to facilitate sex and sectoral analysis.


5. Empirical analysis

The analysis begins with the results of the descriptive statistics, focusing on earningsdifferentials among graduates of the various higher institutions. It moves on tolook at private rates of return and sectoral differences

Empirical results

Table 4 presents the average monthly earnings at each education level for 1995. UsingTable 4, we calculate the earnings differentials between each education level as a

percentage of the average earnings of the lower education level.

Table 4: Mean monthly earnings by educational level by sex and sector (1995)

Variable Education level

Pry Sec NCE Poly Uni

Overall 2196.06 3279.05 3880.31 5330.01 9133.02Male 2365.95 3569.22 3739.96 5182.48 9863.67Female 1611 2734.58 3981.98 5685.71 6637.14Public 2311.87 3530.2 2989.95 4311.86 6528.8Private 2099.95 3067.07 2865.49 4955.15 8966.91

Source: Labour Market Survey, 1995.

A cursory look at Table 4 shows that the mean monthly earnings of workers increasewith more years of schooling. These observations hold true for all categories of workers,whether male, female, public or private. For example, the average monthly earnings forNigerian Certificate of Education graduates for all workers is N3880.37; this is N5330.01and N9133.02 for polytechnic and university graduates, respectively. A comparison ofmale and female earnings reveals a slightly different pattern. The earnings differentialfor males, for example, soars from 4.8% for Nigerian Certificate of Education graduates,to 38.6% and 90.3% for polytechnic and university graduates, respectively. The earningsdifferentials associated with schooling for females stood at 45% for Nigerian Certificateof Education and 42% for the polytechnic. This plummeted to 16.7% for universitygraduates. Also noticeable in Table 4 is that female graduates of NCE and polytechnics,respectively, earn on the average slightly more than their male counterparts with thepercentage difference being in the order of 6.1% and 8.9% for Nigerian Certificate of

19


Education and polytechnic. Whereas a very wide divergence existed in the male–femaleearnings means for university, with the male leading by 48.6 percentage points, the meansof the lower education levels (primary and secondary) also increased with additionalschooling. Noticeable too is that in some cases the mean earnings for secondary graduatesexceed those of their Nigerian Certificate of Education counterparts. The incomedifferentials associated with schooling are reported in Table 5.

Table 5: Monthly earnings differentials associated with schooling

Variable Education level (%)

Pry Sec NCE Poly Uni1 Uni2 Uni3

Overall - 51.2 18.3 62.5 178.5 71.4 135.4Male - 50.9 4.8 45.2 176.4 90.3 163.7Female - 69.8 45.6 107.0 142.7 16.7 66.7Public - 52.7 -15.3 22.1 84.9 51.4 118.4Private - 46.1 -6.6 61.6 192.4 81.0 213.0

Source: Computed by author from Table 4.

1. The column titled Pry (primary) is the difference between the average earnings ofa worker with no education and average earnings of a worker with a primary schooleducation, as a percentage of the average earnings of no education workers.

2. The column titled Sec (secondary) is the difference between average earnings of aworker with primary education and a worker with secondary school education as apercentage of the earnings of the primary school graduate.

3. The column titled NCE is the difference between the average earnings of a workerwith an NCE education and the average worker of secondary school education asa percentage of the average earnings of a secondary school graduate.

4. The column titled Poly is the difference between the average earnings of a workerwith the polytechnic education and those of a worker with secondary schooleducation as a percentage of the average earnings of a secondary school graduate.

5. The column titled Uni1 is the difference between the average earnings of a worker

with university education and the average earnings of a worker with a secondaryeducation as a percentage of the average earnings of secondary school graduates.


with university education and the average earnings of those with a polytechniceducation as a percentage of the average earnings of the secondary school graduates.


with university education and the average earnings of a worker with an NCEeducation as a percentage of the average earnings of secondary school graduates.


The argument that the sector in which a worker is employed affects earnings wasconfirmed from the results in Table 4. A closer look at the table indicates that on theaverage, apart from NCE, the graduates employed in the private sector earned more thantheir counterparts in the public sector, with the difference more pronounced for universitygraduates (37.0%). The NCE graduates employed in the public sector earned slightly(4.2%) more than their counterparts in the private sector. Noticeable too is that earningsdifferentials associated with schooling were moderate for all three levels of education forpublic workers, while they widened for private sector workers (Table 5 and Figure 1).

Presented in Table 6 are the mean earnings by experience groups, and by sex andsector. This enables us to assess the relationship between labour market experience andearnings. A major revelation emerging from a careful consideration of the results in thetable is that mean earnings grow with increased years of labour market experience. Thisfinding holds for all categories of workers, whether male, female, public or private sectorworker, or self-employed. However, assuming that returns to labour market experiencerepresent returns to on-the-job training, it cannot be readily confirmed from the resultsthat the hypothesis of a diminishing marginal returns to increased on-the-job training isvalid.

Table 6: Means earnings by experience group, sex and by sector

Variable Education level (%)

<5 5–9 10–14 15–24 25 >

Overall 4345.93 4710.99 5614.52 6475.11 10,540.64(-)* (8.4%) (19.2%) (15.3%) (62.8%)

Male 4605.25 4908.67 5594.68 7278.14 11,120.11(-) (6.6%) (14.0%) (30.1%) (52.8%)

Female 4000.17 4419.07 5657.08 5077.49 7,904.61(-) (10.5%) (28.0%) (-10.2%) (55.7%)

Public 3462.326 4053.55 4310.75 4760.29 6,893.58(-) (17.1%) (6.3%) (10.4%) (44.8%)

Private 499.36 5007.97 6574.04 7262.24 9,825.54(-) (0.23%) (31.3%) (10.5%) (35.3%)

Self-employed 6083.33 5319.81 - 9259.78 15,986.72(-) (-12.6%) (74.1%) (72.6%)

Note: The values in parentheses are the difference between the average earnings of a worker with a higherlevel of years of experience and the average earnings of a worker with a lower level of years of experience, asa percentage of the average earnings of the less experienced worker.


Results of earnings equations

Table 7 presents the coefficients of the education dummies and experience variablesfor all workers from the earnings equations estimated with ordinary least square

(OLS). All coefficients take the expected signs, except for the coefficients of experienceand experience squares in some cases that are not significant; the model for all workersexplains about 33% of the variations in log earnings. The coefficients on the educationdummies range between 140% and 2.8% (i.e., without the coefficient of constant ofExp2). Table 7 also shows that the coefficient on the education dummy grew with higherlevel of education for all samples. This agrees with the findings of Cohen and House(1994) for Khartoum.

Table 7: Private rate of returns to schooling (all workers)

Coefficient on education dummies

All sample Male Female

(No. 2053) (No. 1392) (No. 661)

Sec. 0.408 (0.053) 0.361 (0.06) 0.54 (0.97)

NCE 0.79 (0.0069) 0.588 (0.09) 0.929 (0.11)

Poly 0.942 (0.066) 0.922 (0.08) 0.929 (0.11)

Uni 1.4097 (0.056) 1.341 (0.067) 1.57 (0.16)

Exp. 0.408 (0.005) 0.035 (0.006) 0.028 (0.011)

Exp2 -0.0003 (0.0001) -0.0004 (0.0001) -0.0002 (0.0002)

Constant 6.83 (0.078) 6.9325 (0.100) 6.68 (0.133)

Adj. R2 0.33 0.31 0.32

F. Stat 165.96 107.18 52.16

Values in parentheses represent the standard deviations of coefficients.

Source: Derived from the estimated regression equation.

A closer examination of the results reveals some differences between men and women,and between public and private sectors. This is shown in tables 8 and 9. In general, it canbe observed that the coefficients, on average, are slightly higher for women than for menin both sectors; similarly, the private sector recorded higher coefficients than the publicsector.

The explanatory power of the regression equation (Adj. R2) stood at 38% and 27%for private and public sectors, respectively, on the average. The coefficient on the educationdummy variables for the public sector ranges between 12% and 34%. The modelperformed better using the statistical test of standard deviation, R2 and F statistics.

Estimates of rates of return accruing to private investment in education in Nigeriaderived from the semi-log earnings functions regression estimates are presented in Table


10. On average, across all levels and types of schooling (lower and higher) and for bothmale and female, schooling yielded about 11% return on the earnings forgone by thehousehold.

Table 8: Coefficient on education dummies (public sector workers)



(No. 896) (No. 579) (No. 317)

Sec. 0.366 (0.081) 0.338 (0.095) 0.533 (0.171)

NCE 0.493 (0.096) 0.509 (0.129) 0.648 (0.180)

Poly 0.777 (0.001) 0.700 (0.122) 1.034 (0.196)

Uni 1.119 (0.887) 1.108 (0.101) 1,219 (0.180)

Exp. 0.016 (0.008) 0.020 (0.009) 0.005 (0.013)

Exp2 -0.0001 (0.0002) -0.0002 (0.00002) -0.0001 (0.0003)

Constant 6.7825 (0.2003) 7.102 (0.162) 7.100 (0.1712)

Adj. R2 0.27 0.2 0.29

F. Stat 52.3 34.2 17.4

Values in parentheses represent the standard deviations of coefficients.Source: Derived from the estimated regression equation.

Table 9: Coefficient on education dummies (private sector workers)



(No. 894) (No. 623) (No. 271)

Sec 0.403 (0.638) 0.358 (0.077) 0.45 (0.114)

NCE 0.660 (0.119) 0.457 (0.16) 0.962 (0.182)

Poly 1.134 (0.087) 0.956 (0.103) 1.48 (0.163)

Uni 1.4097 (0.07) 1.38 (0.081) 1.69 (0.148)

Exp 1.509 (0.005) 0.043 (0.009) 0.064 (0.017)

Exp2 0.0006 (0.00018) -0.0051 (0.0002) -0.001 (0.0004)

Constant 6.653 (0.195) 6.85 (0.13) 6.36 (0.188)

Adj. R2 0.37 0.35 0.42

F. stat 92.14 55.60 39.45

Values in parentheses represent the standard deviations of coefficients.Source: Derived from the estimated regression equation.


The private rate of return as shown in Table 10 increases as the level of educationincreases except for polytechnic graduates, who recorded a 10.7 rate of return, a fall of2% from 12.7% for NCE graduates. The lower rate of return for the polytechnic graduatesmay result because the observation collected for poly does not specify whether somegraduates have the Ordinary National Diploma or the Higher National Diploma Certificate.To illustrate, the rates of return for all samples stood at 12.7% for college of educationgraduates, 10.7% for polytechnic graduates and 16.3% for university graduates.

Table 10: Private rate of return to an additional year of education (%)

Education Total sample Male Female Public Privatelevel

Sec 1.6 -0.5 3.5 1.01 1.4NCE 12.7 7.6 12.96 4.2 8.5Poly 10.7 11.22 8 8.2 14.6Uni 16.7 16.3 10.7 12.6 16.8

Source: Computed from tables 7, 8 and 9 using the rate of return specified in the model and based on theassumption that the years of schooling for primary, secondary, NCE, polytechnic and university are, respectively,6, 12, 15, 17 and 18 years.

In general, the rate of return was quite high for graduates of polytechnics and higherstill for university graduates. This pattern is clearly visible for all the categories of samples– whether public sector, private sector or all samples. The gender differences in the rateof return are also quite evident. Rates of return are higher for male graduates than fortheir female counterparts except for National Certificate of Education graduates, wherefemales recorded a higher rate of return of about 13%. This is contrary to the findings ofPsacharopolous (1994).

Another interesting revelation of this result is that the university and polytechnic asalternative routes after secondary education appear more profitable than colleges ofeducation for males. The low (7.6%) rate of return to college of education after secondaryeducation for males is a clear indication that college of education is not a profitableventure for them. A possible explanation for this pattern of results is thatgraduates of colleges of education are generally employed as teachers in either public orprivate schools, and from experience, teachers are poorly paid in Nigeria. Even years ofexperience do not significantly enhance earnings in the teaching profession, whereastheir secondary school graduate counterparts often find themselves working in othersectors with higher earnings. The reverse is the case for females. As evident in the results,the returns to female NCE graduates in comparison with their secondary schoolcounterparts is significantly high. Thus, college of education as an alternative aftersecondary school is a very profitable venture for females as are university and polytechniceducation.

With regard to sectoral difference, Table 10 tends to support the findings ofPsacharopolous (1994) that the returns in the private/competitive sector of the economyare higher than those of the public/non-competitive sector. As shown in the table, private


sector workers in all education categories earn higher returns than their public sectorcounterparts. This finding supports the use of labour market earnings as a proxy forproductivity in estimating the returns to education.

The argument that earnings are enhanced by the workers’ years of experience waspursued by including a measure of labour market experience, which is often used totrace the productivity enhancing effect of on-the-job training over the life-cycle. It alsoserves as a proxy for seniority, which may in itself lead to higher earnings, but is notnecessarily a guarantee of higher productivity (Cohen and House, 1994). The results inTable 7 show a relatively low coefficient of about 40% for all samples, and the relativecontribution of the experience variables became clearer when they were dropped fromthe model. This specification reduces the overall explanatory power of the regressionequation (adjusted – R2) to about 29% from its 31% with Œ and Œ2 for model 2. Thus,the variation in experience among workers contributes fairly significantly to the inequalityin earnings.


6. Summary

The study was designed to evaluate the nature of returns to higher education inNigeria with a view to accounting for the variations in the rates of return todifferent levels of education. This was accomplished through the use of descriptive

statistics and estimation of the earning functions.

Summary of major findings

The summary of results obtained is thus provided:

• Mean monthly earnings of workers increase with more years of schooling. This wastrue for all categories of workers, whether male, female, public or private workers.

• The coefficients on the dummy variables grow with higher level of education,irrespective of the categories – male, female or all sample.

• The coefficients on average are higher for women than for men. However, the actualrate of returns computed on the basis of the estimated coefficients were higher formale graduates than for their female counterparts in most higher education levelsexcept for polytechnic graduates.

• The private rates of return were low for graduates of colleges of education and verymuch higher for university graduates than for polytechnic graduates.

• On average, across all levels and types of higher education, schooling yielded about11% return on the earnings forgone by the household. Thus higher education can beregarded as a fairly good private investment.

• The return to lower education (primary and secondary) was found to be positivelysignificant.

• Lastly, the results show the means of earnings increasing with higher years of labourmarket experience.

Policy derivatives and conclusion

The significant contribution of education to economic growth and development putsincreased pressure to expand and improve education in most developing countries.

However, most developing countries are faced with the twin problems of economicrecession, with dwindling resources for financing education, and poor educationdevelopment, with deteriorating facilities and infrastructure. Thus, an urgent need has

26


arisen for a reappraisal of both the cost patterns and the financing of education.Experience has shown that in most developing countries like Nigeria, governments

invest heavily in higher education, thereby shouldering an enormous burden. Therefore,a major question that faces policy makers in developing countries is, Which distributionof resources and opportunities for schooling would allow education to have the maximumimpact on development and would best complement investment in physical capital andinfrastructure? Obviously, there are no simple answers. Only the most careful scrutinyof costs in relation to what is achieved will ensure that developing countries get the bestvalue from scarce resources. In addition to seeking ways to reduce the costs of education,many countries are attempting to shift more of the costs from public to private sources.It is in this regard that the findings of this study will be relevant.

This study has found unequivocally that the higher the level of education, the higherthe rate of return to individuals. This makes higher education a worthwhile investmentfor individuals. Thus, more attention should be given to the various methods of costrecovery, including the use of fees for tuition or for meals and accommodation as alreadybeing practised in some institutions in Nigeria, student loans in place of scholarships orgrants, and contributions from employers to help finance vocational education andtraining. These changes will not by themselves solve the financial constraints limitingeducational investment, but they may help governments finance expansion orimprovements that at present cannot be supported because of competing claims on publicfunds.

It should be borne in mind that such shifts from public to private finance may behighly resisted on the grounds of equity. Intuitively, this opposition may not be justifiedand some empirical analysis of the equity implications of alternative investment policieswould be necessary. However, it is logical to reason that fees for tuition or for food andaccommodation, contrary to traditional beliefs, may actually improve the equity ofeducational finance if the public funds saved by this means are then used to increaseselective subsidies for the poor or to increase the provision of education or the quality ofschooling for disadvantaged groups. Likewise, the introduction of student loans in manydeveloping countries could also have a positive distributional impact, since at presentthe high-income students are the ones most likely to benefit from education subsidies.The introduction of student loans could therefore make public funds available for greaterexpansion of primary education, which may well achieve both equity and efficiencyobjectives.

In the sectoral analysis, the magnitude of rates of return in the private sector revealsthat this sector is more competitive and attractive because of the salary package andremuneration offered, which will in turn increase the productivity and efficiency of thesector.

Governments can improve public sector earnings through increase in salary andattractive remuneration, which will induce workers in this sector to concentrate and beproductive. In Nigeria, for example, most government workers (civil servants) engage inpersonal trading and business in order to supplement their earnings, which divides theirinterest and lowers their productivity. Also, the government should encourage more privateinvestors in the economy by providing an enabling environment and good policies for


investment. This will reduce unemployment and encourage investment in education sincegraduates will be assured of ready employment.

Finally, the university, which attracts the highest magnitude of returns, should beproperly funded and well equipped with modern technology, especially the laboratory,library, information system and infrastructure. The private individuals who receive higherearnings as graduates of these institutions should be able to pay at least some part of thisinvestment.


References

Card D. and A.B Krueger.1992a “Does school quality matter? Return to education andcharacteristics of public schools in the United States” Journal of political economy.1– 40, February.

Card D. and A.B Krueger.1992b “School Quality and black-white relative earnings: Adirect assessment” The Quarterly Journal of Economics, 107 (1) February.

Chou E.C. and L.J. Lau. 1978. “Farmer ability and aarm Productivity: A study of farmhouseholds in the Chinagmai Valley, Thailand, 1972–1992”. World Bank DiscussionPapers, No. 62, Washington, D. C.

Cohen, B. and W. J. House. 1994. “Education, experience and earnings in the labourmarket of a developing economy: The case of urban Khartoum”. WorldDevelopment, 22(10): 1549–65.

Chiswick B. R. 1997. “Interpreting the coefficient of schooling in the human capitalearnings function”. World Bank Policy Research Working Paper, No. 179.

Federal Government of Nigeria (FGN). Various years. National Development Plans 1, 2,3 and 4. Federal Ministry of Economic Development, 1970–1990.

Glewwe, P.1991. “Schooling, skills and the returns to government investment ineducation”. Working Paper No. 76 LSMS, Washington, D.C.

Grindling, T. H., M. Goldfarb and C. Chang. 1995. “Changing returns to education inTaiwan: 1978–91”. World Development, 23(2): 343–56.

Heckman, J.1979. “Sample selection bias as a specification error”. Econometrica, 47(1):153–61.

Hossain, S.I. 1976. “Making education in China equitable and efficient”. World BankPolicy Research Working Paper, No. 1814.

Jimenez, E. and J.P. Tan. 1991. “Decentralised and private education. The case of Pakistan”.Comparative Education, 23(2): 173–90.

Krishnan P. 1994. “Family background education and employment in urban Ethiopia”.Centre for the Study of African Economies Working Paper, Series WPS1 94.8.Oxford, England.

Mincer, J. 1974. Schooling Experience and Earnings. New York:: Columbia UniversityPress for National Bureau for Economic Research.

Neuman, S and Ziderman. A. 1991. Vocational schooling, occupational matching andlabour market earnings in Isreal. Journal of Human Resources, 25(2).

Psacharopolous, G. 1985. “Returns to education : A further International Update andImplications”. A World Bank Report Series No. 210: 370. IPSA.

Psacharopolous, G. and E.Velez. 1992. Family productivity, workers benefits and pricechange. A World Bank Publication.

29


Psacharopolous, G. 1994. “Returns to investment in education: A global update”. WorldDevelopment, 22( 9): 132–44.

Psacharopolous, G. and Ng Ying Chu. 1994. “Earnings and education in Latin America:Assessing priorities for schooling investment”. World Bank Policy Research Paper,December.

Schultz, T.W. 1963. The Economic Value of Education. New York: Columbia UniversityPress.

Todaro, M.P. 1982. Economic Development in the Third World. New York: LongmanPublishers.

Vella, F. 1997.“ Estimating Models with sample selection bias: A survey” . vol.33(1).World Bank. 1995. Vietnam Education Financing. World Bank Country Study.

Washington, D.C.


Other recent publications in the AERC Research Papers Series:

The Behaviour of Income Velocity in Tanzania 1967-1994, by Michael O.A. Ndanshau, Research Paper50.

Consequences and Limitations of Recent Fiscal Policy in Côte d’Ivoire, by Kouassy Oussou and BohounBouabre, Research Paper 51.

Effects of Inflation on Ivorian Fiscal Variables: An Econometric Investigation, by Eugene Kouassi,Research Paper 52.

European Economic Integration and the Franc Zone: The Future of the CFA Franc after 1999, Part II,by Allechi M’Bet and Niamkey A. Madeleine, Research Paper 53.

Exchange Rate Policy and Economic Reform in Ethiopia, by Asmerom Kidane, Research Paper 54.The Nigerian Foreign Exchange Market: Possibilities For Convergence in Exchange Rates, by P. Kassey

Garba, Research Paper 55.Mobilizing Domestic Resources for Economic Development in Nigeria: The Role of the Capital Market,

by Fidelis O. Ogwumike and Davidson A. Omole, Research Paper 56.Policy Modelling in Agriculture: Testing the Response of Agriculture to Adjustment Policies in Nigeria,

by Mike Kwanashie, Abdul-Ganiyu Garba and Isaac Ajilima, Research Paper 57.Price and Exchange Rate Dynamics in Kenya: An Empirical Investigation (1970–1993), by Njuguna S.

Ndung’u, Research Paper 58.Exchange Rate Policy and Inflation: The Case of Uganda, by Barbara Mbire, Research Paper 59.Institutional, Traditional and Asset Pricing Characteristics of African Emerging Capital Markets, by Ino

L. Inanga and Chidozie Emenuga, Research Paper 60.Foreign Aid and Economic Performance in Tanzania, by Timothy S. Nyoni, Research Paper 61.Public Spending, Taxation and Deficits: What is the Tanzanian Evidence? by Nehemiah Osoro, Research

Paper 62.Adjustment Programmes and Agricultural Incentives in Sudan: A Comparative Study, by Nasredin A. Hag

Elamin and Elsheikh M. El Mak, Research Paper 63.Intra-industry Trade between Members of the PTA/COMESA Regional Trading Arrangement, by Flora

Mndeme Musonda, Research Paper 64.Fiscal Operations, Money Supply and Inflation in Tanzania, by A.A.L. Kilindo, Research Paper 65.Growth and Foreign Debt: The Ugandan Experience, by Barbara Mbire, Research Paper 66.Productivity of the Nigerian Tax System: 1970–1990, by Ademola Ariyo, Research Paper 67.Potentials for Diversifying Nigeria’s Non-oil Exports to Non-Traditional Markets, by A. Osuntogun, C.C.

Edordu and B.O. Oramah, Research Paper 68.Empirical Studies of Nigeria’s Foreign Exchange Parallel Market II: Speculative Efficiency and Noisy

Trading, by Melvin Ayogu, Research Paper 69.Effects of Budget Deficits on the Current Account Balance in Nigeria: A Simulation Exercise, by Festus

O. Egwaikhide, Research Paper 70.Bank Performance and Supervision in Nigeria: Analysing the Transition to a Deregulated Economy, by

O.O. Sobodu and P.O. Akiode, Research Paper 71.Financial Sector Reforms and Interest Rate Liberalization: The Kenya Experience by R.W. Ngugi and

J.W. Kabubo, Research Paper 72.Local Government Fiscal Operations in Nigeria, by Akpan H. Ekpo and John E.U. Ndebbio, Research

Paper 73.Tax Reform and Revenue Productivity in Ghana, by Newman Kwadwo Kusi, Research Paper 74.Fiscal and Monetary Burden of Tanzania’s Corporate Bodies: The Case of Public Enterprises, by H.P.B.

Moshi, Research Paper 75.Analysis of Factors Affecting the Development of an Emerging Capital Market: The Case of the Ghana

Stock Market, by Kofi A. Osei, Research Paper 76.Ghana: Monetary Targeting and Economic Development, by Cletus K. Dordunoo and Alex Donkor,

Research Paper 77.The Nigerian Economy: Response of Agriculture to Adjustment Policies, by Mike Kwanashie, Isaac

Ajilima and Abdul-Ganiyu Garba, Research Paper 78.


Agricultural Credit Under Economic Liberalization and Islamization in Sudan, by Adam B. Elhiraikaand Sayed A. Ahmed, Research Paper 79.

Study of Data Collection Procedures, by Ademola Ariyo and Adebisi Adeniran, Research Paper 80.Tax Reform and Tax Yield in Malawi, by C. Chipeta, Research Paper 81.Real Exchange Rate Movements and Export Growth: Nigeria, 1960–1990, by Oluremi Ogun, Research

Paper 82.Macroeconomic Implications of Demographic Changes in Kenya, by Gabriel N. Kirori and Jamshed Ali,

Research Paper 83.An Empirical Evaluation of Trade Potential in the Economic Community of West African States, by E.

Olawale Ogunkola, Research Paper 84.Cameroon’s Fiscal Policy and Economic Growth, by Aloysius Ajab Amin, Research Paper 85.Economic Liberalization and Privatization of Agricultural Marketing and Input Supply in Tanzania: A

Case Study of Cashewnuts, by Ngila Mwase, Research Paper 86.Price, Exchange Rate Volatility and Nigeria’s Agricultural Trade Flows: A Dynamic Analysis, by A.A.

Adubi and F. Okunmadewa, Research Paper 87.The Impact of Interest Rate Liberalization on the Corporate Financing Strategies of Quoted Companies

in Nigeria, by Davidson A. Omole and Gabriel O. Falokun, Research Paper 88.The Impact of Government Policy on Macroeconomic Variables, by H.P.B. Moshi and A.A.L. Kilindo,

Research Paper 89.External Debt and Economic Growth in Sub-Saharan African Countries: An Econometric Study, by

Milton A. Iyoha, Research Paper 90.Determinants of Imports in Nigeria: A Dynamic Specification, by Festus O. Egwaikhide, Research Paper

91.Macroeconomic Effects of VAT in Nigeria: A Computable General Equilibrium Analysis, by Prof. D. Olu

Ajakaiye, Research Paper 92.Exchange Rate Policy and Price Determination in Botswana, by Jacob K. Atta, Keith R. Jefferis, Ita

Mannathoko and Pelani Siwawa-Ndai, Research Paper 93.Monetary and Exchange Rate Policy in Kenya, by Njuguna S. Ndung'u, Research Paper 94.Health Seeking Behaviour in the Reform Process for Rural Households: The Case of Mwea Division,

Kirinyaga District, Kenya, by Rose Ngugi, Research Paper 95.Trade Liberalization and Economic Performance of Cameroon and Gabon, by Ernest Bamou, Research

Paper 97.Quality Jobs or Mass Employment, by Kwabia Boateng, Research Paper 98.Real Exchange Rate Price and Agricultural Supply Response in Ethiopia: The Case of Perennial Crops,

by Asmerom Kidane, Research Paper 99.Determinants of Private Investment Behaviour in Ghana, by Yaw Asante, Research Paper 100.An Analysis of the Implementation and Stability of Nigerian Agricultural Policies, 1970–1993, by P.

Kassey Garba, Research Paper 101.Poverty, Growth and Inequality in Nigeria: A Case Study, by Ben E. Aigbokhan, Research Paper 102.Effect of Export Earnings Fluctuations on Capital Formation, by Godwin Akpokodje, Research Paper

103.Nigeria: Towards an Optimal Macroeconomic Management of Public Capital, by Melvin D. Ayogu,

Research Paper 104.International Stock Market Linkages in South Africa, by K.R. Jefferis, C.C. Okeahalam and T.T. Matome,

Research Paper 105.An Empirical Analysis of Interest Rate Spread in Kenya, by Rose W. Ngugi, Research Paper 106.The Parallel Foreign Exchange Market and Macroeconomic Perfromance in Ethiopia, by Derrese

Degefa, Reseach Paper 107.Market Structure, Liberalization and Performance in the Malawi Banking Industry, by Ephraim W.

Chirwa, Research Paper 108.Liberalization of the Foreign Exchange Market in Kenya and the Short-term Capital Flows Problem, by

Njuguna S. Ndung’u, Research Paper 109.External Aid Inflows and the Real Exchange Rate in Ghana, by Harry A. Sackey, Research Paper 110.


Formal and Informal Institutions’Lending Policies and Access to Credit by Small-Scale Enterprises inKenya: An Empirical Assessment, by Rosemary Atieno, Research Paper 111.

Financial Sector Reform, Macroeconomic Instability and the Order of Economic Liberalization: TheEvidence from Nigeria, by Sylvanus I. Ikhide and Abayomi A. Alawode, Research Paper 112.

The Second Economy and Tax Yield in Malawi, by C. Chipeta, Research Paper 113.Promoting Export Diversification in Cameroon: Toward Which Products? by Lydie T. Bamou, Research

Paper 114.Asset Pricing and Information Efficiency of the Ghana Stock Market, by Kofi A. Osei, Research Paper

115.An Examination of the Sources of Economic Growth In Cameroon, by Aloysius Ajab Amin, Research

Paper 116.Trade Liberalization and Technology Acquisition in the Manufacturing Sector: Evidence from Nigeria, by

Ayonrinde Folasade, Research Paper 117.Total Factor Productivity in Kenya: The Links with Trade Policy, by Joseph Onjala, Research Paper 118.Kenya Airways: A Case Study of Privatization, by Samuel Oyieke, Research Paper 119.Determinants of Agricultural Exports: The Case of Cameroon, by Daniel Gbetnkom and Sunday A.

Khan, Research Paper 120.Determinants of Regional Poverty in Uganda, by Francis Okurut, Jonathan Odwee and Asaf Adebua,

Research Paper 122.Exchange Rate Policy and the Parallel Market for Foreign Currency in Burundi, by Janvier D.

Nkurunziza, Research Paper 123.

Structural Adjustment, Poverty and Economic Growth: An Analysis for Kenya, by Jane Kabubo-

Mariara and Tabitha W. Kiriti, Research Paper 124.

Liberalization and Implicit Government Finances in Sierra Leone, by Victor A.B. Davies, Research

Paper 125.

Productivity, Market Structure and Trade Liberalization in Nigeria, by Adeola F. Adenikinju and Louis

N. Chete, Research Paper 126.

Productivity Growth in Nigerian Manufacturing and Its Correlation to Trade Policy Regimes/Indexes

(1962–1985), by Louis N. Chete and Adeola F. Adenikinju, Research Paper 127.

Financial Liberalization and Its Implications for the Domestic Financial System: The Case of Uganda,

by Louis A. Kasekende and Michael Atingi-Ego, Research Paper 128.Public Enterprise Reform in Nigeria:Evidence from the Telecommunications Industry, by Afeikhena

Jerome, Research Paper 129.

Food Security and Child Nutrition Status among Urban Poor Households in Uganda: Implications for

Poverty Alleviation, by Sarah Nakabo-Sswanyana, Research Paper 130.

Tax Reforms and Revenue Mobilization in Kenya, by Moses Kinyanjui Muriithi and Eliud Dismas Moyi,

Research Paper 131.

Wage Determination and the Gender Wage Gap in Kenya: Any Evidence of Gender Discrimination? by

Jane Kabubo-Mariara, Research Paper 132.

Trade Reform and Efficiency in Cameroon’s Manufacturing Industries, by Ousmanou Njikam, Research

Paper 133.

Efficiency of Micro-enterprise in the Nigerian Economy, by Igbekele A. Ajibefun and Adebiyi G.

Daramola, Research Paper 134.

The Impact of Foreign Aid on Public Expenditure: The Case for Kenya, by James Njeru, Research Paper

135.

The Effects of Trade Liberalization of Productive Efficiency: Some Evidence from the Electrical Industry

in Cameroon, by Ousmanou Njikam, Research Paper 136.

How Tied Aid Affects the Cost of Aid Funded Projects in Ghana, by Barfour Osei, Research Paper 137.

Documents

Private returns to higher education in NigeriaThus, the higher the level of education, the higher the rate of return to the individual. Efforts should be made to improve the quality